In silico Design of Porous Polymer Networks: High-Throughput Screening for Methane Storage Materials

Abstract

Porous polymer networks (PPNs) are a class of advanced porous materials that combine the advantages of cheap and stable polymers with the high surface areas and tunable chemistry of metal−organic frameworks. They are of particular interest for gas separation or storage applications, for instance, as methane adsorbents for a vehicular natural gas tank or other portable applications. PPNs are self-assembled from distinct building units; here, we utilize commercially available chemical fragments and two experimentally known synthetic routes to design in silico a large database of synthetically realistic PPN materials. All structures from our database of 18,000 materials have been relaxed with semiempirical electronic structure methods and characterized with Grand-canonical Monte Carlo simulations for methane uptake and deliverable (working) capacity. A number of novel structure−property relationships that govern methane storage performance were identified. The relationships are translated into experimental guidelines to realize the ideal PPN structure. We found that cooperative methane−methane attractions were present in all of the best-performing materials, highlighting the importance of guest interaction in the design of optimal materials for methane storage.

Discovery of optimal zeolites for challenging separations and chemical transformations using predictive materials modeling

“…Selecting optimal candidate materials through predictive modelling is hence a very attractive proposition. Such screening studies have so far focused mostly on single-component adsorption of small, rigid, non-hydrogen-bonding molecules, such as short hydrocarbons [11][12][13][14][15] , carbon dioxide (refs 13,16-19) and hydrogen (refs 13,20). Many of them rely on extrapolation of single-component data 12,[16][17][18] to obtain mixture properties, and others rely predominantly on geometric analysis 13 .…”

Computational development of the nanoporous materials genome

“…One simply needs a single representative SBU to orient on all the nodes in a net, as the net inherently expresses the chirality in its nodes. It is worthy of note that this algorithm (or earlier versions of it) have been used to generate several hypothetical databases of porous materials with a record number of topologies 31,33,63,83,91 . While the sum of these structures consist of roughly 50 topologies, the RCSR contains a total of 2719 nets as of this writing, leaving a large amount of room for further structural diversity 70 .…”

“…Methane storage capacity is by far the most studied sorption property in the field of computational high throughput screening of nanoporous materials [27][28][29][30][31][32][33][34][35][36][37]83 . There are several reasons for this, the first being that there is considerable interest in safely and efficiently storing methane for use as an alternative, clean-burning fuel in motor vehicles.…”

“…Here, computational discovery is typically the first step, always to be followed by experimental work. It would seem obvious to adopt a similar strategy for nanoporous materials design, and sure enough in the past 5 years we have witnessed the proliferation of studies on virtual high-throughput screening of porous materials for gas capture and storage applications [27][28][29][30][31][32][33][34][35][36][37] .…”

Understanding Small‐Molecule Interactions in Metal–Organic Frameworks: Coupling Experiment with Theory

“…Such algorithms have been published for zeolites, [ 139,140 ] and similar tools need to be developed to capture the structure and pore diversity of MOFs. Similarly, data mining and machine-learning approaches [ 141,142 ] will require more MOF-focused classifi ers, including structure and pore geometry, to correlate available simulation and experimental data with possible applications. Hence, tools that do not utilize intensive QM calculations are offering valuable information, even if it's just pore geometry and species-specifi c accessible void space, toward guiding the discovery and application of MOFs.…”